Testing of various types of samples, for example, drinking water, waste water, and biological fluids such as blood and urine, can be performed electrochemically. Assessment of water quality for human health, environmental, and industrial concerns has become increasingly important over the past few decades. Analysis of pollutants, e.g., trace metals in aqueous solution is particularly important because many of these metals (e.g., Hg, Pb) are toxic in low concentrations.
Microelectrodes are useful in analysis of fluids containing electro-active analytes, particularly metals. Microelectrodes can have various geometries, e.g., hemispheres, disks, bands, tubes, rings, and cylinders and generally have one or more dimension on the order of 0.1 to 20 micrometers (Morris, R. B., Franta, D. J. and White, H. S. J. Phys. Chem. 1987, 91, 3559-3564). A microelectrode is an electrode with a dimension (thickness, T or width, W or radius, r) substantially less than the characteristic diffusion length of an analyte of interest. The characteristic diffusion length of an analyte is a function of the duration of the measurement, i.e., it is the square root of the product of the analyte's diffusion coefficient multiplied by the time of the measurement. As is known to those in the art, small metal cations, for example, have typical diffusion lengths which range from about 0.1 to about 1 cm.sup.2 /second, depending, of course, on ionic mobility, among other factors.
As the dimensions of an electrode are made smaller, a number of advantages are gained. The mass transport rate increases, the electrode surface is covered more uniformly, diffusion layer capacitance decreases, the effects of solution resistance decrease, the signal to noise ratio increases, and the need for supporting electrolyte and deoxygenation of the sample is reduced or obviated.
The signal from microelectrodes consists of two components: a faradaic component and a non-faradaic component. The faradaic component represents a chemical reaction occurring on the electrode surface. The non-faradaic component represents the capacitive charging unrelated to the chemistry occurring on the electrode surface. The faradaic component is usually proportional to the periphery of the electrode. The non-faradaic component is proportional to the surface area of the electrode. Electrode geometries which maximize the periphery to surface area ratio also maximize the ratio of the faradaic component to the non-faradaic component and produce readily producible signal. Microelectrodes produce higher periphery to area ratios. The area contacting the sample determines the non-faradaic component, while the periphery determines the faradaic (desired) component.
U.S. Pat. No. 5,120,421 teaches that conventionally sized electrodes often have large uncompensated resistance, making them useless in solutions of low conductivity, e.g., for detecting very low concentrations of analytes.
The diffusion layer (boundary layer), as will be understood by those of ordinary skill in the art, is that volume of fluid sample between where the analyte is at bulk concentration and where the analyte concentration approaches zero (i.e., the fluid layer immediately adjacent to the analyte-covered electrode). It is known to those in the art that the diffusion layer is related to the capacitance of the electrode and the non-faradaic component of the signal, and the diffusion layer can be measured with an electrocapacitance meter (A. Bard and L. R. Faulkner, Electrochemical Methods: Fundamentals and Applications, John Wiley, (1980); and P. T. Kissinger and W. Heineman, Eds., Laboratory Techniques in Electroanalytical Chemistry, 2nd ed., Marcell Dekker, (1990) which are incorporated in their entirety by reference herein). Hence, the diffusion layer acts as a capacitor and any change in applied voltage produces non-faradaic current (component that interferes with the signal). Decreasing the size (dimensions) of the electrode leads to a decrease in the charge stored in the diffusion layer because the electrode surface area is decreased.
The only disadvantage of microelectrodes is the difficulty in measuring low currents. This problem can be overcome by fabrication of an array of identical microelectrodes, so that current signals from multiple microelectrodes can be added together to make a large enough signal for accurate measurements to be made. If the individual electrodes are spaced sufficiently far apart from one another, the currents from each individual electrode are additive and non-interfering.
In recent years many attempts have been made to develop more useful electrode sensors, some with multiple electrodes (arrays) and some with microelectrodes, and methods for making them. U.S. Pat. No. 5,437,999 by Diebold et al. describes an electrode sensor and methods for making such a sensor. The method utilizes photolithographic and screen-printing techniques.
U.S. Pat. No. 5,393,399 by Van den Berg et al. describes an amperometric sensor having a planar structure obtained by photolithographic techniques, useful for measuring the content of an oxygen reducible substance in a fluid.
U.S. Pat. No. 5,670,031 by Hintsche et al. describes an electrochemical sensor with multiple interdigital microelectrodes with structure widths * in the sub-micron range. The spaces between the interdigitated electrodes is about 700 nm, "which are small relative to the distances traveled by the molecules to be detected, in the measuring time."
U.S. Pat. No. 5,217,112 by Almon describes an electrode sensor comprising an auxiliary electrode, a reference electrode and five working electrodes, methods for making such a sensor and voltammetric methods of using such a sensor.
U.S. Pat. No. 5,437,772 by De Castro et al. describes an electrode sensor with multiple interdigitated electrodes which can be coated with mercury forming an array useful for trace metal analysis of samples, especially those containing lead. The sensor is particularly useful in anodic stripping voltammetry techniques.
U.S. Pat. No. 5,103,179 by Thomas et al. describes a water analyzer having several electrodes of the type which normally interfere with one another (active and passive sensors). The water analyzer includes a first electrode (active) which perturbs the sample solution, a second electrode whose reading is affected by the operation of the first electrode, and a sequencing means which activates the first electrode at a time different from the reading of the second electrode.
U.S. Pat. No. 4,874,500 by Madou et al. describes a microelectrochemical electrode structure wherein a monolithic substrate has a well extending into the substrate from the front surface and a passage extending into the substrate from the back surface to the bottom of the well, and an electrode wholly between the front and back surfaces, and a conductor in the passage for electrically communicating the electrode to the back surface.
U.S. Pat. Nos. 5,296,125 and 5,120,421 by Glass et al. describe an electrochemical detection system including a multielement microelectrode array detector capable of acquiring a plurality of signals and electronic means for receiving these signals and converting them into a readout or display providing information about the nature and concentration of elements present in the sample solution.
U.S. Pat. No. 5,676,820 by Wang et al. describes an electrochemical sensor for remote detection, particularly useful for metal contaminants and organic or other compounds. The microelectrode is connected to a long communications cable, allowing convenient measurements of samples as far away as ten to more than 100 feet.
U.S. Pat. No. 5,292,423 by Wang describes a method and apparatus for trace metal testing using mercury-coated screen printed electrodes. Voltammetric and potentiometric stripping analyses are used. Screen printing allows for formation of electrodes with smallest dimensions of about 25 micrometers.
U.S. Pat. No. 5,254,235 by Wu describes a microelectrode array, contructed from a woven a minigrid, preferably comprising multiple microdisks. The minigrid is formed by a plurality of vertical and horizontal conductive filaments woven together, wherein the horizontal elements are parallel to the exposed surface of the substrate and the vertical elements extend parallel to the longitudinal axis of the electrode. The microelectrode array is formed by potting the minigrid in an electrically insulating material, such as epoxy resin. The ends of the vertical elements are exposed at the mesuring end surface by grinding away an outer layer of the substrate such that the end-surface and vertical filament ends define a measuring surface for the microelectrode. The filaments are preferably cylindrical. The typical thickness of the minigrids range from about 3 micrometers to about 6 micrometers. The spacing between the microdisks may be selected to minimize adverse effects resulting from overlapping diffusion layer areas surrounding each of the microdisks.
None of the above references, nor any reference known to the present inventors, describes an array of microband electrodes which gives true steady-state current. That is, there has been no apparatus or method for making an apparatus having an array of microband electrodes with adequate dimensions and spacing such that the individual microband electrodes do not interfere with one another.
As is understood by those of ordinary skill in the art and as detailed in U.S. Provisional Application Ser. No. 60/030,319, spherical (disks) and hemispherical microelectrodes exhibit chronoamperometric steady-state behavior in relatively short periods of time. Semi-infinite planar electrodes (i.e., those with sufficiently large surface areas that edge effects are negligible) exhibit Cottrellian chronoamperometric behavior: the current approaches zero at long times (not steady-state, as the current would be with a microdisk).
Diffusion to microdisks and semi-infinite planar electrodes is not uniform. The flux of analyte to the electrode surface occurs primarily at the outer circumference and edges, respectively. The central portion of the electrode is relatively free of analyte while the edges contain the bulk of the deposited analyte. This behavior has prompted the design and fabrication of small ring electrodes (microelectrodes) which exhibit chronoamperometric behavior similar to that of microdisks (steady-state), but with improved signal to noise ratios (Tallman, D. E. Anal. Chem., 1994, 66:557). These electrodes suffer from limits on the thickness of the ring (about 2 microns) based on the wavelength of the light source used in the photolithography used to make them.
An alternative to ring microelectrodes is a band microelectrode (microelectrodes which are rectangular or square in shape).
The prior art teaches that the amperometric behavior of microband electrodes is quasi-steady-state. The limiting current, it has been understood by those in the art, decays as a function of time but not a fast as true Cottrellian behavior (as does an infinite planar electrode), but it doesn't reach a steady-state (like a hemisphere or disk microelectrode) either.
U.S. Pat. No. 5,254,235 by Wu, mentioned above, explains that the failure to obtain true steady-state microelectrode arrays "comes from the fact that either the individual electrodes themselves exhibit only virtual steady-state current because of their size, such as microband electrodes, or the disks are randomly dispersed with separations therebetween that are too small causing current shielding." (Virtual steady-state is equivalent to quasi-steady-state.)
Morris, R. B., Franta, D. J. and White, H. S. in J. Phys. Chem., 1987, 91:3559 describe a band microelectrode in which the thickness of the band can be as thin as 200 angstroms, made by either thermal or electron beam evaporation sources and a crystal monitor to control the thickness of the deposited film. The microband electrodes in this reference were made of platinum and gold. The width (W) of the microelectrodes was between 0.5 and 1 cm, and the thickness (T) between 20 and 500 angstroms. As the authors of the reference explain, "since the gap width of any break in the band electrode is insignificant compared with the length of any one segment, and since each segment is in electrical contact with the bulk surface film, the overall geometry approximates that of a single band of known macroscopic length."
Early attempts to describe the diffusion to microband electrodes used a hemicylindrical geometry to describe the edge of the band (Wehmeyer, K. R., Deakin, M. R., and Wightman, R. M. Anal. Chem., 1985, 57:1913; Kovach, P. M., Caudill, W. L., Peters, D. G. and Wightman, R. M. J. Electroanal. Chem., 1985, 185:285). These two references describe microband electrodes with dimensions of 5-2300 nanometers thickness by (macroscopic) centimeters width, and about 18 microns by (macroscopic) about 0.32 centimeters width, respectively. Both of these papers teach that microbands exhibit quasi-steady-state amperometric behavior.
All of the above mentioned microband electrodes have small widths but very much larger, macroscopic lengths. The advantages of the band geometry result from the high magnitude of the current because of the long length, coupled with high mass transport diffusion rate because of the very small width.
The theory regarding microband behavior, i.e. that microbands exhibit quasi-steady state behavior at long times, is based on the assumption that the width of the band is macroscopic. Thus, at long times, the behavior of a microband electrode is dominated by semi-infinite planar diffusion.
Thormann, W., van den Bosch, P., and Bond, A. M. (Anal. Chem., 1985, 57:2764) fabricated an array of Au microband structures in which both dimensions were in the microelectrode regime (T=0.1 .mu.m, W=15 .mu.m). The spacing of these electrodes, however, was only 30 .mu.m apart. This results in a .theta. value of 0.75 which could lead to overlap of individual diffusion layers at short periods of time. This would manifest itself as a semi-infinite diffusion response and therefore the limiting current would not achieve a true steady state. Additionally, tests on these electrodes involved cyclic voltammograms only and not any chronoamperometric measurements.